Data Scientist ｜Global Internet Business

【About the Company】

The company plans, develops and operates a social shopping site with over 8.5 million members (as of the end of January 2021) in 164 countries.
It is a marketplace where you can purchase fashion items from all over the world from personal shoppers (sellers) from all over the world. You can find items that are sold out or hard to find in Japan, brands that have not yet landed in Japan, and other products that are special and highly valuable to each customer. The English version of the site is steadily expanding its user base, especially in North America. We will continue to aim to become a marketplace that can be used by people around the world.

【Job Description】 

You will be responsible for maximizing the value of BUYMA's services using data science.
This is a position where you will be able to work on everything from the development of machine learning models and algorithms to the development of machine learning systems for services (using data obtained through the operation of BUYMA and Stylehouse), depending on your experience and wishes.

【Specific Projects】
・Similar image search
・Name collation and normalisation of item IDs on display
・Recommendation algorithm development
・Personalisation of search result order
・Fraud detection algorithm development
・Content generation and interactive UX using generative AI

◆Department: Data Technology Group
The following members belong to the group that develops and operates the data, machine learning and search infrastructure. In this position, you will be part of the Data Scientist team.
Search and MLOps engineers: 2
Data scientist: 1
Data engineers: 2
*Data analysis and effectiveness measurement is mainly handled by data analysts in other departments, you will also work with them. In this position, you will be responsible for areas closer to engineering.

【Technology stack】
Language: Python 3.X SQL, Shell Script
DB: BigQuery
BI: Looker, Redash
Code management: GitLab
Communication: slack, zoom, esa, redmine
Machine learning infrastructure: Kubeflow PIpelines, Vertex AI
# You can choose your own analytical environment if required

【Working conditions & treatment】

Flextime System

・Saturday/Sunday/National Holiday
・Annual Paid Leave
・New Year Holidays
・Congratulations & Condolence Leave
・Refresh Leave

・Commuting allowance
・Full social insurance
・Use of recreational facilities and sports facilities of Kanto IT Software Health Insurance Association is available
・Subsidy system for babysitter use (for internal events)
※The following system is also available for engineers※
・Lending notebook PC of your choice when you join the company
・Lending of dual displays if desired
・Subsidy system for purchasing a smart phone

In-house system
・Award system (MVP, etc., for each personnel evaluation)
・Encouragement of seminar participation
・Overseas business trip system
・Book purchase subsidy system

【Required】
・Experience of programming in Python with the ability to analyse and create predictive models
・Knowledge of and strong interest in machine learning such as regression, classification and clustering
・Knowledge of basic statistical analysis, such as understanding correlation and causality, and various tests.
・Ability to communicate analysis results using machine learning and statistical analysis methods in an easy-to-understand manner.
・Ability to construct hypotheses, propose and promote measures based on an understanding of services and systems

【Preferred】
・Experience in using machine learning to actually improve and grow services.
・Desire to work with marketing teams and other engineers to solve problems
・Ability to choose the right analysis method for the task, balancing accuracy with time cost etc.
・Experience of participating in data analysis competitions such as Kaggle
・Knowledge of mathematical optimisation algorithms and experience in their implementation
・Willingness to learn from books, papers, etc. on your own if necessary.
・Experience of using SQL to extract data